The average miner on p2pool (and thus the average share variance) is ~1.5GH/s You "could" do it as a 1 GH/s miner but you already have above average variance so I wouldn't.  So a miner w/ 3GH/s @ 1,500 would have the same variance as the avg miner does @ target 750 (current min share diff). A miner w/ 15 GH/s @ 7,500 would have the same variance as a 1.5GH/s miner @ target 750 (current min share diff). While I don't suggest someone raise it that high the more hash power you have the more you can raise diff and still have better than avg variance. |
|
|
|
|
Dual relays are pretty much useless for anyone with a farm. I have 8 going on 12 rigs. giga has 32 rigs.
|
|
|
|
I was thinking more like a maintained optimized iso. Burn and boot. Like LinuxCoin (except updated sometime in the last year) or like BAMT but based on x64 and focused on cgminer. BTW I use kano guide for my rigs (and used BAMT before I needed 8 GPU support). Kano guide works and is invaluable but it is 30 something steps the first of which involves burning a CD to make the linux environment to make the USB key. Not exactly "plug and play". |
|
|
|
But you can purchase that ethernet relay board, and it's completely open, so what's the prob on using them? :O
Nothing. One could also write their own miner or build their own pool too. I would imagine not everyone would want to research boards, find one that is compatible, cut custom wiring harness, and code up an API. Still there is nothing that is stopping them. Having a common API is the reason I am interested in making a standardized board and API. I already have a board and it gives me manual remote power control. However to expand beyond that I need there to be enough similar boards so that third parties look to support them via API. Like: * Having a cgminer monitoring program (like anubis) which reboots any rig in a farm as needed when a GPU crashes. * Building a hearbeat into BAMT (and as a package for other linux distros) that the relay board listens for and reboots a server when its haearbeat stops. * paying a third party to provide SMS support for the relay board so that you can get messages when a rig is rebooted. There are advantages of having large number of people using the same hardware and software. |
|
|
|
I really don't understand how some of you simply ignore the fact that transaction fees are irrelevant in comparison to block reward, and will be for the forseeable future. Even if they weren't, there are plenty of people out there with the incentive to disrupt the Bitcoin network by mining without including transactions, regardless of what the fees are.
Nothing you have suggested so far changes that.
Not sure what your definition for "forseable future" is. As long as tx fees are essentially 0 so no matter what the block reward is ~0/anything is still essentially 0%. Even a modest rise in fees can provide meaningful change to that dynamic. Hypothetical scenario. Over next 12 months tx volume doubles to ~100 per block and block reward falls to 25 BTC. If miners start enforcing fees and avg fee ends up being just 0.01 BTC per tx that is 1 BTC in fees and 25 BTC reward. Fees are a small but hardly insignificant 4% of total revenue in just 12 months.Hopefully on a longer time frame say 3 or 4 years tx volume will quintuple (or more). At say 250 tx per block and 12.5 BTC subsidy. An avg fee of just 0.01 BTC per tx results in fees making up 16% of total revenue. If USD:BTC remains the same then mystery faces falling revenue and will need to adapt (or make less). If USD:BTC prices then global hashing power will also rise as miners chase larger block rewards (including fees) and "mystery" makes up a smaller portion of the network. Going forward bitcoin will need to be supported more and more by fees anyways. Any "problem" myster presents is temporary at best. Breaking changes to protocol to "fix" a temporary problem seem dubious. |
|
|
|
|
Nobody has used proof of work to build a consensus in a distributed network (currency or otherwise). That is kinda the cornerstone of Bitcoin. The rest is pretty routine p2p functionality stuff (node detection, data relaying ,etc).
|
|
|
|
You shouldn't lose any hashes. BTC mining requires negibible bandwidth. We are talking KB per second. Messing w/ HT link can cause system instabilities though. You likely will find more power savings turning off junk you don't need. On my MSI 890FXA-GD70 I have USB 3.0 - disabled Secondary LAN - disabled EIDE controller -disabled SATA controller - disabled SATA RAID - disabled on board sound - disabled serial port - disabled monitor LED ( wtf  ) - disabled I have Sempron single core underclocked to 1.6Ghz and undervolted (? can't remember to what). It likely could go lower but I just tried 1.6GHz and dropped voltage 2 steps and it worked fine. Never pushed it beyond that. Oh and the always helpful Power after AC Loss: Always On |
|
|
|
How's this project going?
My prototype is working. I will share some pics tonight. I have been busy working on watercooling stuff. The code used to drive the prototype is ugly. I am still debating if I will commercialize it. On one hand it likely will be a lot of work with low margins. On the other hand if enough units are sold that will likely create the demand for third party applications to provide support which helps me. I'll post the photos and see what kind of feedback I get. |
|
|
|
I made change, pls see my previous post.
Yeah code looks good. Thanks for supporting p2pool I am sure smaller miners appreciate the reduced min difficulty (and variance) that will cause. |
|
|
|
bitcoind has always priced transactions per 1000 bytes (SI kB). I can certainly change it, but I don't think upstream will accept it that way.
No reason to change it then. The warning limit here is 0.00025 BTC per byte, or 0.25 BTC per 1000 bytes. Surely that is sufficiently high? Oops. No that is more than sufficiently high. If the "spam fee" is higher than the user-defined fee, the spam fee is left alone. Otherwise, the user fee overrides it. Gotcha. I see now that prior to this segment the min "spam fee" is already computed based on coinage of the inputs. Looks good to me. Thanks for the prompt response. |
|
|
|
Frans Lategan, who will be one of the expert speakers at the annual ITWeb Security Summit, in May, says he will reveal for the first time at the Summit newly-discovered weaknesses in the gold standard cryptography. Describing the vulnerabilities as “scratches in the paintwork, rather than a train smash”, Lategan says his findings nevertheless indicate that vulnerabilities can exist even in trusted algorithms in use to protect currency as valuable and widespread as Bitcoin. Ok so it is only academic. Obviosuly one wouldn't wait 2+ months to release findings on a flaw unless it is minor ... Also I love the " protect currency as valuable and widespread as Bitcoin." Lategan explains that Bitcoin, a fast-growing global digital currency that resides solely in the cloud, has already been the victim of attacks. How exactly does he know Bitcoin has been a victim? Unless he is talking about things like DDOS and thefts which have nothing to do with the vulnerability? Hmm... Either he is full of shit trying to pump up his presentation ahead of the conference or his is the single most unethical cryptographer on the planet. "I know of a vulnerability which is costing other money and undermining public trust in cryptography so I will wait for two months before telling anyone about it ... er I will tell me about it, just not what it is." |
|
|
|
As I understand you suggest to enter ".../1500+1" or more ?
I have rigs with 4 cards.
Your advice is very valuable.
I'll follow you at once.
Yes. It isn't required. It is simply a way to transfer more variance (but same avg payout) to you and help smaller miners. Think of it as a non-financial donation. While it will take you "longer" (say 4 minutes vs 2 minutes) to find shares that is easily handled by someone with a lot of hashing power. Once you modify the your miner usernames to user/1500+1 you will see something like this in p2pool window. New work for worker! Difficulty: 1.00000 Share difficulty: 1500.0000344 Total block value 50.00045000 BTC
If you want to change the difficulty in the future (higher or lower) you just need to change the miner's username. If you want to go back to dyanmic minimum difficulty (currently ~750) just remove the /1500 |
|
|
|
TL/DR version:
If BTC cost is 1000x higher than VISA cost then BTC is dead now, so uninstall the client. VISA gross profit margin is 28%. Thus the cost to users is ~4x their cost. If BTC cost is 1000x higher then BTC ultimate cost to users would be 250x what VISA cost is. Kinda hard to attract users to that. The reality is BTC cost is a tiny fraction of VISA once the network is "fully grown".
i think mining will become cost ineffective for the average joe long before that. i imagine if bitcoin were to grow to these levels, corporate mining will come into play, and most likely exchanges and particularly bitcoin payment processor companies will have a large portion of the mining infrastructure for additional profit, fast proliferation of transactions, and security. I agree but I mean for USERS. If the math was right and BTC cost was 1000x VISA cost and visa has a 4x markup then BTC price to end users (non-miners) would be 250x as high as VISA even if miners worked for cost. That is obviously untrue. |
|
|
|
|
My understanding is it is to identify the underlying encryption protocol. All keys today have the prefix 0x04 because they all use ECDSA (w/ secp256k1 curve). It is possible for Bitcoin to support future algorithms (if ECDSA became compromised or degraded). The 0x04 allows the client to "check" the algorithm being used. If it finds a key w/ anything other than 0x04 it will fail but future clients could be extended.
You have to imagine if ECDSA was "partially compromised" existing keys will remain for a long time. Having a encryption identifier would allow a client to handle both encryption methods at the same time. Old clients would be ECDSA only and new clients would support the old & new algorithms.
If you mean why 0x04 not 0x03 or 0x99? I don't think it has any specific meaning other than an identifier.
|
|
|
|
It was self-important enough to create an untraceable identity. I am sure it assigned the same importance to its Bitcoins as well. I doubt it. The early days of Bitcoin was mostly an academic experiment. Even the 10K BTC pizza didn't happen until years later when Bitcoin at least had gained the potential for future adoption. I doubt early wallets were well protected. I don't even think Satoshi believed Bitcoin (the version he released) would go this far. Everything about the paper, early writing, launch indicates it was more a "proof of concept". Then again if one of those early mining rewards ever gets moved/consolidated then we will know for sure. |
|
|
|
if satoshi is human and humans make mistakes, i think it's possible a large number of his private keys have already been lost permanently.
I've considered that. However, some have speculated quant/s had to be involved. Could this really be the work of one man? Perhaps this group knew exactly what was going to happen. Under such a scenario, such carelessness could not have happened. The coins have not been spent. They might have a purpose. The code has been found to be meticulous. Um the concept is amazing. Beyond amazing. Satoshi (individual or group) solved a large number of problems considered unsolvable and put into place a lot of elements which provide future security and enhancement. However the code (translating that amazing concept into lines of code) is not meticulous. It is buggy, poorly documented, often times poorly implemented. The version you see now has involved tens of thousands of hours of work by post Satoshi developers and still is pretty "ugly" as codebases of that size go. |
|
|
|
I received another donation for p2pool miners. I'll distribute it sometime in the next day or two.
Greetings from ...XVD (27GH/s) + my 5 BTC donation to growth p2pool. BTW is all ok with my getwork? Something improve? I use now 38 VTX 7970, plan increase to 64 within 2 weeks. Panda Mouse. Awesome Panda Mouse. With 27GH/s you may wish to consider a higher difficulty target. It lowers difficulty for smaller miners. You can alter difficulty by using a miner username in this format. username/[difficulty target]+[local diff] difficulty target is how "hard" of a share you look for (compensation is based on value of shares not just nominal number) local diff is optional but it is the diff your miners see. I set it to "1" which allows tracking stats in cgminer across miners and time. currently the p2pool min difficulty is ~750. For a small miner that may mean no share for hours. With 27GH/s @ 750 difficulty you will find 1 share every 2 minutes. Increasing difficulty to say 1500 would still mean you find 1 share (worth twice as much) every 4 minutes and it helps to keep min diff lower for smaller miners. I have ~15 GH/s and use a higher diff of 1500 BTW. Glad to see another large miners using p2pool. |
|
|
|
i was looking into it, but itd have to be a installable os, as you cant upgrade drivers with persistance the normal way
not too sure how bamt is allowing it but meh :/
You can upgrade drivers with persistence. Although driver changes are so rare it would simply make more sense to release a new version of the OS as needed. |
|
|
|
|
Yeah the math is off because it is based on VISA scale volume and 8x increase in BTC price.
While BTC may someday be that large it likely will be when subisidies are much lower. fees will rise but not as much as subsidies are declining. While the nominal value of BTC in fiat may rise in BTC terms future block likely will be worth much less than 50 BTC. We will see the first step of that in 2013. Block subsidy will decline to 25 and while miners may push for higher fees it is unlikely going to be even 10% of the subsidy drop.
So the network is likely to be cheaper and more efficient in the future (on a per tx basis) if/when BTC hits even Paypal level volumes much less VISA. Also the "math" fails to account this is just an article on VISA newest datacenter. They have lots of datacenters all over the world increasing their infrastructure cost by a magnitude.
TL/DR version:
If BTC cost is 1000x higher than VISA cost then BTC is dead now, so uninstall the client. VISA gross profit margin is 28%. Thus the price to users is ~4x their cost. If BTC cost is 1000x higher then BTC ultimate price to users would be 250x what VISA price is. Of course I don't think BTC will ever cost more than a tiny fraction of VISA but it shows how silly the 1000x number is.
|
|
|
|
|
sorry to hear that. 0.5 added a "feature" called mother which when a card locks up it reboots the server and disables OC on that card. It is posisble that it what happened.
I use cgminer and really just use BAMT for the OS. I find it more stable and easier to configure that way.
Still I am considering offering a bounty for a cgminer only OS as BAMT doesn't support x64 and thus is limited to 6 or 7 GPUs per rig.
|
|
|
|
|
Show Posts
